Garuda - Garba Rujukan Digital

KOMPARASI ALGORITMA NAIVE BAYES DAN DECISION TREE PADA KLASIFIKASI PENERIMAAN PESERTA DIDIK BARU Aprilia, Kesi; Khaira, Ulfa; Ferdian Hutabarat, Benedika
Simtek : jurnal sistem informasi dan teknik komputer Vol. 10 No. 2 (2025): Oktober 2025
Publisher : STMIK Catur Sakti Kendari

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51876/simtek.v10i2.1694

Penerimaan Peserta Didik Baru (PPDB) merupakan tahapan penting dalam menyeleksi calon peserta didik yang memenuhi kriteria lembaga pendidikan. Di Yayasan Sahabat Qur’an Al-Karim Jambi, peningkatan jumlah pendaftar tidak sebanding dengan kapasitas penerimaan, sehingga dibutuhkan proses seleksi yang lebih tepat dan objektif. Penelitian ini bertujuan untuk mengaplikasikan metode data mining dengan melakukan perbandingan antara algoritma Naïve Bayes dan Decision Tree CART untuk mengetahui algoritma yang paling efektif dalam klasifikasi penerimaan peserta didik baru. Proses penelitian mencakup tahap praproses data, pembagian dataset dengan proporsi 80% untuk pelatihan dan 20% untuk pengujian, penerapan kedua algoritma, serta dilakukan evaluasi mengggunakan k-fold cross validation dan confusion matrix. Temuan penelitian menunjukkan bahwa algoritma Decision Tree memberikan performa paling unggul dengan akurasi 98,77%, presisi 99,19%, recall 97,56%, dan f1-score 98,34%, sedangkan algoritma Naïve Bayes memperoleh akurasi 94,44%, presisi 96,54%, recall 89,02%, dan f1-score 92,04%.

Komparasi Metode Naive Bayes dan K-Nearest Neighbors Terhadap Analisis Sentimen Pengguna Aplikasi Zenius Abdillah, Tegar; Khaira, Ulfa; Hutabarat, Benedika Ferdian
Jurnal PROCESSOR Vol 19 No 1 (2024): Jurnal Processor
Publisher : LPPM Universitas Dinamika Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33998/processor.2024.19.1.1596

The purpose of this research is to compare the performance of Naive Bayes and K-Nearest Neighbor (KNN) methods in analyzing user sentiment on the Zenius application. The evaluation is done by checking the precision, precision, recall, and F1-Score scores of both methods as well as visualizing the results of sentiment analysis with one of the methods used. The advantage of this research is a deeper understanding of how Naive Bayes and KNN techniques work in sentiment analysis in the context of the Zenius app. Furthermore, this research aims to evaluate the performance results of two techniques, Naive Bayes and KNN, in sentiment analysis. From the results of testing split data scenarios using Split Validation with training data and testing data 90:10. Naive Bayes accuracy reached 88.41%, while KNN reached 100%. In this study, KNN outperformed Naive Bayes in terms of precision, recall, and F1-Score values. The results of data visualization show that the direction of the sentiment generated tends to be positive. This study not only provides a deeper understanding of the performance of Naive Bayes and KNN techniques in sentiment analysis for the Zenius application, but also provides a comprehensive evaluation of their performance. This research is expected to serve as a guide for developing more effective sentiment analysis methods for similar applications in the future.

RANCANG BANGUN APLIKASI ANDROID UNTUK KLASIFIKASI PENYAKIT TANAMAN JERUK MENGGUNAKAN CNN DAN METODE EXTREME PROGRAMMING Rio Gilang, Divo; Ulfa Khaira; Benedika Ferdian Hutabarat
Jurnal INSTEK (Informatika Sains dan Teknologi) Vol 10 No 2 (2025): OCTOBER
Publisher : Department of Informatics Engineering, Faculty of Science and Technology, Universitas Islam Negeri Alauddin, Makassar, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/instek.v10i2.61172

Jeruk merupakan salah satu komoditas pertanian penting yang memiliki nilai ekonomi tinggi dan permintaan pasar yang besar. Namun, produktivitas dan kualitas jeruk sering terancam oleh penyakit pada kulit, seperti Citrus Bacterial Spot, Citrus Canker, dan Huanglongbing (HLB), yang dapat menurunkan nilai jual dan daya saing jeruk lokal. Deteksi penyakit secara manual memerlukan waktu dan keahlian khusus, sehingga dibutuhkan solusi berbasis teknologi yang cepat dan akurat. Penelitian ini bertujuan merancang dan membangun aplikasi mobile untuk identifikasi penyakit kulit jeruk menggunakan metode Extreme Programming (XP) serta mengevaluasi fungsionalitas sistem. Model klasifikasi citra dibangun menggunakan arsitektur MobileNetV2. Keunikan riset ini terletak pada integrasi lightweight deep learning model dengan pendekatan XP, sehingga menghasilkan aplikasi yang tidak hanya akurat, tetapi juga efisien dan siap diterapkan langsung di perangkat mobile dengan sumber daya terbatas. Evaluasi sistem menunjukkan akurasi validasi yang tinggi serta efisiensi kinerja berdasarkan metrik penggunaan sumber daya perangkat tanpa memaparkan detail teknis. Aplikasi ini mampu mengidentifikasi penyakit kulit jeruk secara cepat dan akurat, sehingga berpotensi membantu petani dalam pengambilan keputusan pengelolaan tanaman.

KLASTERISASI WILAYAH RAWAN KRIMINALITAS DI KOTA JAMBI (2022-2024) MENGGUNAKAN ALGORITMA K-MEANS Sulistina, Sri; Eko Prasetyo Utomo, Pradita; Ferdian Hutabarat , Benedika
Jurnal INSTEK (Informatika Sains dan Teknologi) Vol 10 No 2 (2025): OCTOBER
Publisher : Department of Informatics Engineering, Faculty of Science and Technology, Universitas Islam Negeri Alauddin, Makassar, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/instek.v10i2.62051

Penelitian ini bertujuan mengelompokkan kerawanan kriminalitas di 11 kecamatan Kota Jambi pada periode 2022–2024 menggunakan algoritma K-Means. Data kriminalitas diperoleh dari Polresta Kota Jambi, sedangkan data demografis (jumlah penduduk, luas wilayah, dan kepadatan penduduk) diperoleh dari BPS Kota Jambi. Seluruh variabel numerik dinormalisasi menggunakan metode min–max. Penentuan jumlah klaster dievaluasi pada K = 2–10 menggunakan Elbow Method , Silhouette Coefficient, dan Davies Bouldin Index (DBI). Pada tahun 2022 dan 2023, Elbow menunjukkan titik siku pada K = 3 (masing-masing bernilai 0,872 dan 1,124), dengan nilai Silhouette maksimum pada K = 3 (0,476 dan 0,466), sedangkan DBI mencapai nilai minimum pada K = 9 (0,197 dan 0,3963). Pada tahun 2024, Elbow kembali mengarah ke K = 3 (1,182), Silhouette tertinggi diperoleh pada K = 5 (0,536) dengan nilai yang masih kompetitif pada K = 3 (0,492), dan DBI kembali minimum pada K = 9 (0,2458). Dengan mempertimbangkan konsistensi Elbow dan Silhouette serta kemudahan interpretasi, dipilih K = 3 sebagai jumlah klaster optimal. Evaluasi konsistensi menggunakan Rand Index menunjukkan nilai 1 sepanjang periode, sehingga menguatkan bahwa tiga klaster merupakan struktur pengelompokan yang stabil.

Application of You Only Look Once (YOLO) Method for Sign Language Identification Reni Triyaningsih; Pradita Eko Prasetyo Utomo; Benedika Ferdian Hutabarat
Jurnal Nasional Teknik Elektro dan Teknologi Informasi Vol 14 No 4: November 2025
Publisher : This journal is published by the Department of Electrical and Information Engineering, Faculty of Engineering, Universitas Gadjah Mada.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/jnteti.v14i4.21931

Limited understanding of sign language has widened the social gap for deaf people, creating barriers in communication and social interaction. To address this challenge, technology-based solutions are required to facilitate inclusive communication. Deep learning-based detection methods, particularly the You Only Look Once (YOLO) algorithm, have gained attention for their speed and accuracy in real-time object detection. This research aims to develop and evaluate a YOLO training model for the identification of Indonesian sign language system (sistem isyarat bahasa Indonesia, SIBI). The dataset was obtained from resource person at the State Special School Prof. Dr. Sri Soedewi Masjchun Sofwan, SH. Jambi, and enriched with additional images collected from external subjects. Augmentation techniques with Roboflow were applied to expand the dataset, and several training schemes were implemented. Model performance was assessed using confusion matrix while considering accuracy and indications of overfitting. The results showed that the quality and quantity of training data, as well as the epoch values, strongly influenced the accuracy of the trained model. The best performance was achieved with 40 primary images per label class, augmented to 60 images, and trained over 24 epochs, resulting in a confusion matrix accuracy of 99.9%. The implemented model was able to recognize SIBI gestures in real-time using a webcam with fast processing. Overall, the proposed YOLO-based model successfully identifies sign language in real-time and demonstrates strong potential for reducing communication barriers among deaf people. However, further refinement and expansion of the dataset are recommended to improve effectiveness and enable broader real-world applications.

Sosialisasi Dan Workshop Aplikasi Berikan-Jambi Sebagai Media Binaan Manajemen Konservasi Ikan Lubuk Larangan Guci Emas Pada Desa Muara Pijoan Sukmono, Tedjo; Utomo, Pradita Eko Prasetyo; Wulandari, Tia; Hutabarat, Benedika Ferdian; Suprayogi, Dawam
Jurnal Pengabdian Masyarakat Pinang Masak Vol. 6 No. 2 (2025): Vol 6 No 2 (2025) Desember 2025
Publisher : Fakultas Sains dan Teknologi, Universitas Jambi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22437/jpm.v6i2.51619

In community life, cultural elements play a crucial role as guidelines for social order. The culture applied in society extends beyond objects to behavior, attitudes, attitudes toward nature, and even the management of nature itself. One of Jambi Province's abundant natural resources is its fish biodiversity. The Batanghari River has a significant impact on both fish diversity and culture. This cultural connection to fish remains deeply embedded in the community to this day. Furthermore, fish folklore is also found, which is beneficial for fish conservation. Ethnoscience is not merely a cultural knowledge; it is also an intangible heritage of Jambi Province. Therefore, ethnoscience and fish biodiversity are interconnected. Ethnoscience and fish biodiversity are now integrated into a single information system, the Jambi Fish Biodiversity Information System (BERIKAN JAMBI) https://ikanjambi.unja.ac.id/. This integrated information system contains information about the fish found in the Batanghari River. The socialization and application workshop given to the Guci Emas awareness group in Muaro Pijoan Village aims to provide knowledge and understanding regarding the Berikan Jambi information system in supporting ethnoscience studies of the community in the area and as a conservation step through the fostered group, so that it can make it easier for the community to obtain information about local folklore regarding certain fish.

PEMBERDAYAAN KELOMPOK SADAR WISATA GUCI EMAS MUARO PIJOAN MELALUI IMPLEMENTASI SISTEM INFORMASI DESA WISATA Utomo, Pradita Eko Prasetyo; Sukmono, Tedjo; Wulandari, Tia; Hutabarat, Benedika Ferdian; Suprayogi, Dawam
MIMBAR INTEGRITAS : Jurnal Pengabdian Vol 5 No 1 (2026): JANUARI 2026
Publisher : Biro Administrasi dan Akademik

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36841/mimbarintegritas.v5i1.7640

Desa Muaro Pijoan di Kabupaten Muaro Jambi memiliki potensi wisata berbasis kearifan lokal melalui kawasan Lubuk Larangan Guci Emas, namun promosi digitalnya masih terbatas. Kegiatan pengabdian ini bertujuan mengimplementasikan sistem informasi desa wisata berbasis web dan media sosial untuk memperkuat strategi promosi serta meningkatkan kapasitas digital Kelompok Sadar Wisata (Pokdarwis) Guci Emas. Metode pelaksanaan menggunakan pendekatan partisipatif melalui pelatihan literasi digital, pendampingan pembuatan website, serta produksi konten promosi digital. Hasil kegiatan meliputi terbentuknya website https://lubukguciemas.com, aktivasi akun media sosial resmi, modul pelatihan, serta peningkatan kemampuan mitra dalam mengelola konten secara mandiri. Program ini terbukti efektif dalam memperkuat promosi wisata berbasis teknologi dan kearifan lokal, sekaligus menjadi model replikasi pengembangan desa wisata digital di Provinsi Jambi.

Pemanfaatan Sistem Presensi Online untuk Peningkatan Kinerja Pegawai Klinik Razi, Muhammad; Suratno, Tri; Hutabarat, Benedika Ferdian; Setiawan, Dedy; Lestari, Dewi; Noverina, Yolla; Alfajri, Willy Bima
Cahaya Pengabdian Vol. 2 No. 2 (2025): Desember 2025
Publisher : Apik Cahaya Ilmu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61971/cp.v2i2.244

This community service activity was implemented at Praktek Dokter Pattimura, which has three branches in Jambi City. The main issue faced by the partner was the manual employee attendance system using physical books or unconnected fingerprint devices, leading to slow, inaccurate, and high-risk data loss records, making it difficult for management to monitor employee discipline. This condition increased the administrative burden and hindered the effectiveness of human resource management. The solution offered was the development of a digital attendance application based on web and mobile platforms. This system utilizes face recognition technology for identity verification and GPS integration to ensure attendance can only be done within the clinic area. The system includes a mobile application for daily employee attendance and a web dashboard for owners and administrators to monitor real-time presence data. This solution makes the attendance recording process faster, more accurate, transparent, and integrated, thereby encouraging increased employee discipline and clinic management effectiveness. The implementation was carried out through the Prototype model, including coordination, system development, training, and continuous assistance.

Application of You Only Look Once (YOLO) Method for Sign Language Identification Reni Triyaningsih; Pradita Eko Prasetyo Utomo; Benedika Ferdian Hutabarat
Jurnal Nasional Teknik Elektro dan Teknologi Informasi Vol 14 No 4: November 2025
Publisher : This journal is published by the Department of Electrical and Information Engineering, Faculty of Engineering, Universitas Gadjah Mada.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/jnteti.v14i4.21931

Limited understanding of sign language has widened the social gap for deaf people, creating barriers in communication and social interaction. To address this challenge, technology-based solutions are required to facilitate inclusive communication. Deep learning-based detection methods, particularly the You Only Look Once (YOLO) algorithm, have gained attention for their speed and accuracy in real-time object detection. This research aims to develop and evaluate a YOLO training model for the identification of Indonesian sign language system (sistem isyarat bahasa Indonesia, SIBI). The dataset was obtained from resource person at the State Special School Prof. Dr. Sri Soedewi Masjchun Sofwan, SH. Jambi, and enriched with additional images collected from external subjects. Augmentation techniques with Roboflow were applied to expand the dataset, and several training schemes were implemented. Model performance was assessed using confusion matrix while considering accuracy and indications of overfitting. The results showed that the quality and quantity of training data, as well as the epoch values, strongly influenced the accuracy of the trained model. The best performance was achieved with 40 primary images per label class, augmented to 60 images, and trained over 24 epochs, resulting in a confusion matrix accuracy of 99.9%. The implemented model was able to recognize SIBI gestures in real-time using a webcam with fast processing. Overall, the proposed YOLO-based model successfully identifies sign language in real-time and demonstrates strong potential for reducing communication barriers among deaf people. However, further refinement and expansion of the dataset are recommended to improve effectiveness and enable broader real-world applications.

Title

Found 29 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 29 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 29 Documents
Search